New build BSOD's multiple times a day

[transientmetal]

Specs: Windows 8.1 64bit
Graphics:GeForce GTX 980
MOBO:MSI Z170 Krait Gaming
Ram: AMD radeon DDR4 16 gb (8x2)
CPU: Intel I5 6600k

Issue:
I built this computer a few weeks ago and it crashes whenever I play a video game or even browse the web. BlueScreenDisplay confirms the problem is a BSOD even though I don't actually see it; the entire computer freezes (whether playing a game or browsing the web), I hear 3 rapid beeps and the computer hard restarts with no additional display or message. The restart is completely normal (no error messages) as if I had just turned on the computer.

I've monitored the temperatures for both CPU and GPU and nothing is out of the ordinary. I've run Furmark 3 times, 15 minutes each run - the GPU maxes out at 73 C with a constant 80 fps. I've done a CPU stress test through the Intel utility with temperature maxing out around 44 C. A memtest was run for about 4 hours and 0 errors were found.I'm testing each individual stick today for longer to really make sure the ram is not the issue. I've disabled all audio drivers other than what my headset uses. I've scanned the computer with MalwareBytes as well as Avira antivirus (no issues). WhoCrashed reads a total of 30 crash dumps, with only 3 being related to something other than windows drivers (ntoskrnl.exe, ks.sys). Those 3 are:
cfosspeed6.sys (cFosSpeed Driver, cFos Software GmbH)
sshid.sys (SteelSeries HID Driver, SteelSeries ApS)
nvlddmkm.sys (NVIDIA Windows Kernel Mode Driver, Version 359.00 , NVIDIA Corporation)

The crashes related to these 3 drivers happened relatively early and I updated them as best I could. These crashes have not reoccurred after the first. The most common crashes seem related to ntoskrnl.exe. I'm hesitant to run DriverVerifier as I don't want to lose my files in case I have to reinstall windows. I'm hoping an answer can be found to my issues before I resort to that as it is extremely frustrating to not know what the problem is.

the Minidump files

 

[johnbl]

third bugcheck: nvidia graphic driver, the stack was corrupted.
this can happen if the nvidia card overheats, has poor power but it can also happen when you have audio conflict between a audio device and the graphics card audio support for HDMI. I would turn off any unused audio sources, reset the BIOS to default, this will make the BIOS rescan the hardware and re assign the various interrupts and DMA channels. It will pass a small database to windows which will use it to configure remaining devices. You can also go into device manager and disable any un -used high definition audio devices. I am thinking of graphics audio devices when you monitor does not get a sound source from the video cable.
You can disable the motherboard audio in BIOS if you don't use it.

it is hard to tell what overwrote the stack with a minidump. it would show up in a full or kernel memory dump.
you can also get corrupted stacks like this with malware and bad network drivers. just depend on the order that the drivers are stored in memory. My guess is two audio drivers using the same direct memory access channel and responding to the same request. You can disable the different audio sources, or maybe put a speaker on all of the audio outputs and you will hear a nasty sounds when there is a conflict, generaly a BIOS reset or update helps.

--------
second bugcheck was caused by a bad memory address passed to a USB audio device.
This can be also caused by the overclocking software, or a bug in the USB support for the BIOS, a bug in the USB drivers for the USB port, or a bug in the actual drivers for the USB audio device.
generally BIOS versions and USB driver versions after 2013 work pretty well. I would remove the overclocking software, update the BIOS and maybe run a memtest86 test to confirm you don't have a memory setting problem.
(the BIOS update will add microcode for your processor and it will update the default memory timings for your motherboard)



--------
looking at the most current bugcheck first and working backward.
-------------
bugcheck caused by a bad memory address used by nvidia streaming software
remove: overclocking software:
NTIOLib_X64.sys Mon Nov 26 02:11:53 2012
this driver is just too old for your new CPU, the old driver will force incorrectly high voltages on your new lower power CPU.
IE wrong voltage for the clock frequency used.

I think you are having some sleep issues also, you might want to update the BIOS to the current version and install the current drivers for your motherboard. take a look at these (confirm it is the correct board and version of windows)
http://us.msi.com/product/motherboard/support/Z170-KRAIT-GAMING.html#down-driver&Win8 64

machine info:
BIOS Release Date 07/15/2015
Manufacturer MSI
Product Z170 KRAIT GAMING (MS-7984)
Version 2.0
Processor Version Intel(R) Core(TM) i5-6600K CPU @ 3.50GHz
Processor Voltage 8bh - 1.1V
External Clock 100MHz
Max Speed 3500MHz
Current Speed 3500MHz

 

[transientmetal]

thank you for the quick reply John! I was using MSI Live Update 6 to update the drivers and I just noticed that it wasn't actually downloading the MSI BIOS files. I used the link you gave me to flash the bios but it seems that file only updated my bios to version B.30 while the Live Update states that a newer version is available (B.40).Unfortunately I'm encountering the same problem that I can't download through the Live Update app. There is no mention of a newer build on the MSI website Not only that, but I found the file NTIOLib_X64.sys in the Live Update folder as well. Should I just remove Live Update altogether?

Despite updating the bios, the computer still crashed with a DRIVER_IRQL_NOT_LESS_OR_EQUAL error. However, the BSOD was actually visible this time. I'm going to reset the bios and follow your other steps. Hopefully it will fix the issue!

 

[transientmetal]

So unfortunately the computer still BSOD's on a regular basis and I still haven't figured out what the problem is. To make matters worse, the computer doesn't even restart properly after a BSOD; the screen says collecting data, goes to 100% and then hangs there forever unless I manually restart it by holding the power key. A normal restart doesn't even work anymore; the computer turns off and then turns back on but there is no power to the display, the keyboard or the mouse, forcing me to a do a hard restart. I've removed MSI Live Update and run various driver tools to find and update any old drivers but no old drivers are detected. In fact, the Intel Driver Utility doesn't even detect my current drivers! WhoCrashed and BlueScreenView don't help in identifying the problem as most of the BSOD messages are generic in nature. The most recent minidump files are linked below

Minidump Files

I'm more than willing to provide any other information that might help me resolve this issues. I'm willing to upgrade to windows 10 as well if Windows 8 is the problem.

 

[johnbl]

- you might underclock your 2133MHz memory to see if you can prevent the memory errors.
make sure the memory is correctly set to the proper timings the manufacture number was not specified in the BIOS settings. just the 2133MHz speed. (confirm memtest86 runs without errors)


-looks like all the system chipset drivers should be updated:
http://us.msi.com/product/motherboard/support/Z170-KRAIT-GAMING.html#down-driver&Win10 64


looks like you have a second set of overclocking software installed.
Intel Extreme Tuning Utility Performance Tuning Driver
iocbios2.sys Thu Apr 09 22:42:49 2015

Run memtest86 on its own boot image and confirm your memory is ok,
if no problems found, you need to remove the overclocking driver and see if you still get a bugcheck.

generally a single bit corruption like this would be a problem in the actual hardware, BIOS, or hardware specific drivers for your system. The bugcheck was pretty fast (53 mins)
the system loaded a system file from storage and it was corrupted by the time it was stored in RAM (1 bit corruption)

you might want to check with the motherboard vendor to see if there is a known issue with the actual motherboard
Manufacturer MSI
Product Z170 KRAIT GAMING (MS-7984)
Version 2.0

or a bug in the intel chipset drivers for the z170 chipset.
I would be looking to update these drivers before running hardware tests:
I2cHkBurn.sys Mon Feb 17 17:34:54 2014 (5302B8BE)
iaLPSS2_GPIO2.sys Mon May 04 05:41:36 2015 (55476900)
iaLPSS2_I2C.sys Mon May 04 05:41:19 2015 (554768EF)
iaLPSS2_UART2.sys Mon May 04 05:41:32 2015 (554768FC)
ICCWDT ICCWDT.sys Tue Apr 21 02:51:44 2015 (55361DB0)

if you can not find the cause of the problem and you think your hardware is not at fault,
then start cmd.exe as an admin and run
verifier.exe /standard /all

then reboot your machine, this will force verifier to look for common bugs in device drivers and force bugcheck if it finds a programming mistake. The bad driver will be named in the memory dump.
note: use verifier.exe /reset
to turn off verifier functions when done testing.






looking at the most current bugcheck, you had a 1 bit memory corruption.
the system wrote binary 11100110 but read binary 11000110
a single bit was modified, the system detected the change and called a bugcheck.